The Future of Games
1 Introduction
Lots of things are changing in the games industry
We’re going to talk about a few of them Hardware trends Software trends Game design trends Business model trends Hardware Trends More cores for all platforms
Multithreaded, job-based game architecture is increasingly important Burst processing ECS! Increase in functional programming models to solve some multithreaded issues Map-Reduce Immutable data Heterogeneous parallel programming
GPUs are different from CPUs Local memory (VRAM) Different instruction set Highly parallel Throughput optimised Batch up computations, submit to GPU, get results later Increasingly suitable for applications other than rendering Physics Machine learning CPUs are latency optimised Designed for frequent branching Computation results are available “immediately” Cloud
Computation as a service Lowers the barriers to online services Great for small developers Scalability Can spin up new servers to handle load on the service Pay as you go Popular platforms Amazon Web Services (AWS) Microsoft Azure Google Cloud Platform (GCP) Google App Engine Increasingly used for hosting dedicated game servers Mobile
Mobile hardware is getting close in performance to current consoles Already past the last generation of consoles Some new input mechanics Touch screen Accelerometer GPS (Ingress - Pokemon Go, Harry Potter) No D-pad on an iPhone Examples: iPhone iPad 3DS Controllers
Wiimote PS Move Xbox Kinect – oh dear Wii U gamepad – oh dear Nintendo Switch - Labo VR controllers Haptic controllers Companion apps Touchscreen Accelerometer GPS Camera Stereoscopic 3D
Nintendo 3DS – but is it on the way out? More games support 3D – but still nowhere near all Consoles are capable of making games in 3D, but graphics horsepower isn’t really there VR and AR
Oculus Rift HTC Vive PSVR Google Glass HoloLens Magic Leap Longshots
Recent things – how did they pan out?
Valve’s console – failure! Ouya: Android console – failure! Big support on Kickstarter OnLive / Playstation Now – success! AppleTV as a console – working! Google TV – failure!
Another longshot Death of the console industry, or one of its players In the twilight of Moore’s Law, the transitions to multicore processors, GPU computing, and HaaS cloud computing are not separate trends, but aspects of a single trend – mainstream computers from desktops to ‘smartphones’ are being permanently transformed into heterogeneous supercomputer clusters. Henceforth, a single compute-intensive application will need to harness different kinds of cores, in immense numbers, to get its job done.
The free lunch is over. Now welcome to the hardware jungle.
Herb Sutter 2012 http://herbsutter.com/welcome-to-the-jungle/ Software Trends Rendering
Lighting models Physics-based lighting now common Real-time raytracing – Nvidia RTX Quantitative improvements – “4K ready” Procedural content Physics
Finite element methods Destructible worlds Very few games have pursued the “construction physics” approach of Red Faction: Guerilla Crackdown 3 maybe Soft bodies Fluids Continuous collision detection Quantitative improvements Animation
Parametric blending – Uncharted 4 Improvements in motion capture - Hellblade Facial animation Automatic phoneme detection Autonomous control Natural motion Motion planning Artificial Intelligence
Machine learning Planning Motion Navigation Action Less scripting, more 'brain' Game Design Trends Casual games
Social gaming Facebook gaming Play with friends Strange relationship with the rest of the games industry Mobile Instant-on, bite-sized gaming Play in a few minutes On transit, waiting in line, etc Input differences dramatically influence game design Candy Crush, Angry Birds, etc. Board game ports have been successful on the iPad Telemetry and Metrics
Publishers and developers can get data on what players are doing Either in development, or after launch Focus development effort on things players do most or where they get stuck Release new versions or updates based on telemetry data
Vital for some business models like social and free-to- play Compulsion loops
“Kompu Gacha” – banned in Japan in 2012
The dark side of telemetry and metrics
If you optimise for certain metrics, you can get games that do well at those metrics, but very little else
Loot boxes – a good or bad thing? In some European countries, possibly an illegal thing? User-generated content
Photo modes, video capture Important enough that current-gen consoles have intrinsic support for it Photos actually look good enough to be worth sharing ;) SteamWorks: end-to-end support for creating and distributing mods and assets
Can actually make money from it Level editors, sometimes even on console games
Halo Forge, DOOM SnapMap Super Mario Maker, Little Big Planet
Keeps players coming back without ongoing cost to developer Remasters and spiritual successors
Remasters from earlier console generations (sometimes much earlier) are popular Needs skilled conversion work – BluePoint Games Often 'just' a visual upgrade – people want the game to feel the same, but to look modern
Spiritual successors – devs who worked on one IP make something very similar Pillars Of Eternity, Yooka-Laylee, Bioshock, Axiom Verge, Obduction, Mighty No. 9 Reliant on nostalgia and/or disdain for modern design Often crowdfunded Can self-limit potential market Might just crash and burn (ironically, movie tie-in games have somewhat died out) Business Model Trends Crowdfunding / transparent development
A few high-profile games on Kickstarter – Elite: Dangerous – We Happy Few – Star Citizen – Prison Architect Pillars Of Eternity Steam Greenlight / Early Access Early access means you are at least guaranteed to get some kind of game to play immediately Customers want to influence design, want transparency Unreal Tournament – development halted (thanks Fortnite) Hits-driven business
More money going to top titles, less to everyone else Games have always been this way, but the gap is getting wider Mobile changed this only temporarily The same may happen with VR The top titles are probably happy with this, but over the long term it may not be great for the industry Film has smaller “Indie” successes, but they’re rare in games It’s not all doom and gloom though Cheap/free engines lower the barrier to entry for the small guy Easy access to large distribution channels (App Store etc) Crowdfunding Battle for the living room
Netflix on console – sure, but TVs often integrate it too. Lots of buzz around Bandersnatch. AppleTV as a console – not a major player? Google Chromecast – Android games Valve's console – didn't take off
This is first generation where we've had “.5” consoles Designed to promote 4K visuals Games must use extra power strictly for visuals, not gameplay Can make users of the “.0” hardware feel inferior Reviewers will be encouraged to concentrate on the X / Pro versions Free to play
AEM Acquire, Engage, Monetize Advertisement-funded Microtransactions Free for the main game, but pay for extras Pay to progress Or at least to progress quickly Pay to win Pay for better gear than other players Once the economy of the game is ruined, start up a new server More common in Asia Subscriptions
Holy Grail for publishers Predictable long term income
MMOs on PC have been doing this for a long time e.g. World of Warcraft
Hardware platforms Xbox Live Gold PlayStation Plus Users can play multiplayer, get free games Recoups the cost of running servers Content monetization
Paid DLC Paid virtual swag Skins Hats Taunt animations If you buy a taunt animation and kill me and I see it, that means the animation must exist on my machine too ;) Small amounts from many players add up DLC isn't free to develop
Taking a cut of player transactions Digital distribution
Tower Records has gone out of business Music sales have mostly moved online Amazon iTunes Apple Music
Blockbuster has gone out of business DVD sales and rentals have mostly moved online Amazon iTunes Netflix
GameStop is going out of business...? Game “first sales” have mostly moved online Not yet...
GameStop is still a big player in the games industry Publishers want to move to online, but don’t want to upset GameStop, Walmart, etc
Advantages to digital Lower cost of goods (basically free to distribute online) Stops used game sales – which is where a lot of store revenue is Low-friction purchasing
Some clear successes already iTunes App Store Steam, especially its compulsive promotions Mobile
Very good device penetration Don’t need a console when you already have a phone Cheaper games Price point seems to be $0.99 or free Much harder to market games Can’t pay for shelf space
iPhone vs DS, PSP, Vita Rise and fall of Zynga
Social Game Developer Primarily Facebook games Free to play with microtransactions Strong focus on telemetry Very successful IPO Social was the new moneymaker Facebook rule changes have hurt Zynga Can’t spam friends as much as before Zynga’s business is still hits-based Farmville was huge Cityville less so Farmville 2 was a mild success Huge (75%) drop in stock price in 2012 CEO named one of the worst CEOs of 2012 E-Sports
Rapidly growing business 2015 revenue: Asia $321 million US $224 million Europe $172 million Rest of world $29 million
Tournament prizes Game Voyeurism
Free marketing Twitch TV Games are starting to offer Twitch integration Rise Of The Tomb Raider allows Twitch viewers to select between difficulty mods for the streaming player
YouTube YouTube personalities are sometimes trusted more than mainstream reviewers - “no filter” Getting your game mentioned by Yogscast or PewDiePie can spike sales Summary
Lots of stuff happening in the industry
Some trends are very clear: Multiprocessor hardware Digital distribution
Some less clear, susceptible to laws / player takeup: Social games VR / AR